An intelligent routing method applied to cross-domain communication networks

Abstract

The present invention discloses an intelligent routing method applied to a cross-domain communication network, which is applied to a cross-domain communication network including multiple communication nodes in at least two spatial domains. The method comprises: determining each link of the cross-domain communication network at the current moment, and determining the current link weight corresponding to each link based on the link transceiver information of each link at the current moment; wherein a link represents a direct communication relationship between two communication nodes; constructing a current weighted network graph corresponding to the cross-domain communication network based on each communication node, each link and the corresponding current link weight; and determining the route from each communication node to each other communication node in the cross-domain communication network at the current moment based on the current weighted network graph. For such a complex communication scenario of the cross-domain communication network, the route with the highest transmission efficiency can be determined, thereby improving the overall data transmission efficiency of the cross-domain communication network.

Description

Technical Field

[0001] The present invention relates to the field of computer processing technology, and in particular to an intelligent routing method applied to a cross-domain communication network. Background Art

[0002] In recent years, cross-domain communication scenarios have continued to emerge. Currently, the Ad hoc On-Demand Distance Vector Routing (AODV) protocol is commonly used as a routing protocol in cross-domain communication networks. This routing protocol mainly selects the route with the least hops as the optimal choice.

[0003] However, in actual cross-domain communication scenarios, due to the different data transmission rates of different links, sometimes data packets are transmitted faster through links that reach through multiple hops than through direct paths with fewer hops. The above routing method cannot determine the route with the highest transmission efficiency for cross-domain communication networks. Summary of the Invention

[0004] The present invention provides an intelligent routing method applied to a cross-domain communication network, which can determine the route with the highest transmission efficiency in complex communication scenarios of the cross-domain communication network and improve the overall data transmission efficiency of the cross-domain communication network.

[0005] The present invention provides an intelligent routing method applied to a cross-domain communication network, wherein the cross-domain communication network includes multiple communication nodes in at least two spatial domains, including:

[0006] Determine each link of the cross-domain communication network at the current moment, and determine a current link weight corresponding to each link based on link transceiver information of each link at the current moment; wherein the link indicates that there is a direct communication relationship between two communication nodes;

[0007] Constructing a current weighted network graph corresponding to the cross-domain communication network based on each of the communication nodes, each of the links, and the corresponding current link weights;

[0008] Based on the current weighted network graph, a route from each communication node to each other communication node in the cross-domain communication network at the current moment is determined; wherein the route includes at least one link.

[0009] Optionally, the current link weight corresponding to each link is determined based on the link receiving and transmitting information of each link at the current moment, including: for each link, based on the link receiving and transmitting information of the link at the current moment, determining the communication quality evaluation attribute and the link priority attribute corresponding to the link; based on the communication quality evaluation attribute and the link priority attribute, determining the current link weight corresponding to the link.

[0010] Optionally, the link receiving and sending information includes: data packet flow information flowing through the link, the time required for the link to transmit data packets, link bandwidth information, link transmission delay and link packet loss rate. The link receiving and sending information of the link at the current moment is used to determine the communication quality assessment attributes and link priority attributes corresponding to the link, including: determining the communication quality assessment attributes corresponding to the link based on the link flow information, the time required for the link to transmit data packets and the link bandwidth information corresponding to the link; and determining the link priority attributes corresponding to the link based on the link transmission delay, link packet loss rate and preset link level information.

[0011] Optionally, the determining of the communication quality assessment attribute corresponding to the link based on the link traffic information corresponding to the link, the time required for the link to transmit a data packet, and the link bandwidth information includes: determining the average traffic information based on the link traffic information corresponding to the link and the time required for the link to transmit a data packet; determining the link load information corresponding to the link based on the average traffic information and the link bandwidth information; and determining the communication quality assessment attribute corresponding to the link based on the link load information.

[0012] Optionally, determining the communication quality evaluation attribute corresponding to the link based on the link load information includes: performing an exponential operation on the link load information to obtain the communication quality evaluation attribute corresponding to the link.

[0013] Optionally, the link priority attribute corresponding to the link is determined based on the link transmission delay, the link packet loss rate and the preset link level information, including: taking the inverse of the preset link level information and performing an exponential operation to obtain an importance level reference value corresponding to the link; performing a weighted sum operation on the link transmission delay, the link packet loss rate and the importance level reference value to obtain the link priority attribute corresponding to the link.

[0014] Optionally, determining the current link weight corresponding to the link based on the communication quality assessment attribute and the link priority attribute includes: inputting the communication quality assessment attribute and the link priority attribute corresponding to the link into a pre-trained weight prediction model to obtain the current link weight value corresponding to the link.

[0015] Optionally, based on each of the communication nodes, each of the links and the corresponding current link weights, a current weighted network graph corresponding to the cross-domain communication network is constructed, including: taking each of the communication nodes in the cross-domain communication network as a node, each of the links as a connecting edge, and the current link weight corresponding to each of the links as the weight value of the corresponding connecting edge, to construct a current weighted network graph corresponding to the cross-domain communication network.

[0016] Optionally, based on the current weighted network graph, determining the route from each communication node to each other communication node in the cross-domain communication network at the current moment includes: for each node in the current weighted network graph, determining each node except the first node as the second node; taking the first node as the starting node and any second node as the ending node, using a preset path optimization algorithm to determine the optimal path between the first node and the second node; determining the first communication node corresponding to the first node, the target communication node corresponding to the second node, and each target link corresponding to the optimal path; and determining the path formed by each target link as the route from the first communication node to the target communication node in the cross-domain communication network at the current moment.

[0017] Optionally, the at least two spatial domains include an airspace domain, a surface domain and an underwater domain, and the multiple communication nodes include at least one airspace communication node, at least one surface domain communication node and at least one underwater domain communication node; the link between the airspace communication node and the airspace communication node corresponds to a radio communication mode; the link between the airspace communication node and the surface domain communication node corresponds to a radio communication mode; the link between the airspace communication node and the underwater domain communication node corresponds to a laser communication mode; the link between the surface domain communication node and the surface domain communication node corresponds to a radio communication mode; the link between the surface domain communication node and the underwater domain communication node corresponds to an underwater acoustic communication mode; the link between the underwater domain communication node and the underwater domain communication node corresponds to an underwater acoustic communication mode.

[0018] The technical solution of the embodiment of the present invention is applied to a cross-domain communication network including multiple communication nodes in at least two spatial domains. The method includes: determining each link of the cross-domain communication network at the current moment, and determining the current link weight corresponding to each link based on the link transceiver information of each link at the current moment, wherein the link indicates that there is a direct communication relationship between two communication nodes, and then, based on each communication node, each link and the corresponding current link weight, constructing a current weighted network graph corresponding to the cross-domain communication network, thereby determining the route from each communication node to each other communication node in the cross-domain communication network at the current moment based on the current weighted network graph, wherein the route includes at least one link. The technical solution provided by the embodiment of the present invention, for such a complex communication scenario of the cross-domain communication network, can dynamically adjust the link weight in real time according to the link transceiver information of different links in the cross-domain communication network, thereby determining the route corresponding to the current moment according to the adjusted link weight, and can determine the route with the highest transmission efficiency, thereby improving the overall data transmission efficiency of the cross-domain communication network.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 This is a flow chart of an intelligent routing method applied to a cross-domain communication network according to the first embodiment of the present invention;

[0022] Figure 2 is a schematic diagram of a current weighted network graph corresponding to a cross-domain communication network according to an embodiment of the present invention;

[0023] Figure 3 This is a flow chart of an intelligent routing method applied to a cross-domain communication network according to a second embodiment of the present invention;

[0024] Figure 4 This is an overall flow chart of an intelligent routing method applied to a cross-domain communication network provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "water surface domain", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Before introducing this solution, the application scenario of this solution can be introduced first. This embodiment is applicable to any situation where it is necessary to determine the routing between communication nodes in a cross-domain communication network.

[0028] A cross-domain communication network is a network composed of communication nodes in different spatial domains. A cross-domain communication network includes multiple communication nodes in at least two spatial domains. A communication node is a connection point in a cross-domain communication network. It can be a redistribution point or a communication endpoint, such as a server, computer, drone, or robot.

[0029] More specifically, at least two spatial domains may include airspace, surface domain, and underwater domain. Multiple communication nodes include at least one airspace communication node, at least one surface domain communication node, and at least one underwater domain communication node. That is, it can be understood that the cross-domain communication network includes multiple airspace communication nodes corresponding to the airspace (for example, drones flying in mid-air), multiple surface domain communication nodes working in the surface domain (for example, servers configured on a ship), and multiple underwater domain communication nodes working in the underwater domain (for example, servers configured in an underwater environment, underwater robots, anchor points in the sea, etc.).

[0030] Routing plays a vital role in communications. It ensures efficient and reliable network transmission by selecting the optimal path to deliver data packets from source to destination. Currently, routing strategies based on traditional routing protocols are unable to achieve efficient and reliable transmission in communication scenarios across air, surface, and underwater domains. In addition, current routing determination strategies only update routes based on static information when a communication node is disconnected, and do not perform timely dynamic optimization based on the actual dynamic communication information in the cross-domain communication network. To address the above situation, this paper proposes a method for dynamically and adaptively determining link weighted attributes and then performing intelligent routing based on the link weighted attributes. This method can determine the link with the highest transmission efficiency in complex communication scenarios across air, surface, and underwater domains, thereby achieving the effect of intelligent routing.

[0031] Example 1

[0032] Figure 1 This is a flow chart of an intelligent routing method applied to a cross-domain communication network provided by the first embodiment of the present invention. Figure 1 As shown, the intelligent routing method applied to the cross-domain communication network includes:

[0033] S110: Determine each link of the cross-domain communication network at the current moment, and determine a current link weight corresponding to each link based on link transceiver information of each link at the current moment.

[0034] A link refers to a communication line from one communication node to an adjacent communication node in data communication, with no other switching nodes between the two communication nodes. In this embodiment, a link indicates that there is a direct communication relationship between the two communication nodes.

[0035] For a cross-domain communication network composed of communication nodes in multiple spatial domains such as airspace, surface, and underwater, a single homogeneous communication method is difficult to meet the data transmission requirements of the cross-domain communication network. This is mainly reflected in the following aspects: the radio communication method uses electromagnetic waves to transmit information, which can achieve high coverage and high transmission rate communication in the air, but electromagnetic waves attenuate greatly in water, so it is difficult to meet the requirements of long-distance communication transmission across the surface or underwater; the underwater acoustic communication method uses sound wave transmission, and the transmission loss of water waves in water is relatively small. It is the main means of underwater wireless long-distance communication transmission, but the communication rate is very low; the laser communication method uses laser as a carrier for data transmission and is suitable for cross-surface communication. Based on this, in this embodiment, the cross-domain communication network includes multiple types of communication nodes, namely, airspace communication nodes, surface domain communication nodes, and underwater domain communication nodes, and there are differences in the link communication methods between different types of communication nodes. Specifically, the link between airspace communication nodes corresponds to the radio communication mode; the link between airspace communication nodes and surface communication nodes corresponds to the radio communication mode; the link between airspace communication nodes and underwater communication nodes corresponds to the laser communication mode; the link between surface communication nodes and surface communication nodes corresponds to the radio communication mode; the link between surface communication nodes and underwater communication nodes corresponds to the underwater acoustic communication mode; the link between underwater communication nodes and underwater communication nodes corresponds to the underwater acoustic communication mode.

[0036] Among them, the link transceiver information is the communication information of a link in the process of receiving and / or sending data packets. For example, for a certain link, the link transceiver information may include: the data packet flow information flowing through the link, the time information required for the link to transmit data packets, link bandwidth information, link transmission delay, and link packet loss rate, etc. The link weight is used to characterize the priority of a link. The link weight can also be understood as the negative impact value of each link. The path finally selected is the path with the smallest sum of negative impact values ​​from the starting node to the destination node. Therefore, the larger the weight of a link, the less likely it is to be selected and the lower the priority. The current link weight is the link weight determined based on the link transceiver information corresponding to a link at the current moment.

[0037] Specifically, during the operation of the cross-domain communication network, the communication relationship between each communication node can change in real time, that is, the links and the number of links included in the cross-domain communication network also change in real time. In this embodiment, according to a pre-set timed task, at a preset time interval, it can be determined which communication nodes in the cross-domain communication network are in a state of direct communication at the current moment. The communication connection relationship between these communication nodes is the various links of the cross-domain communication network at the current moment.

[0038] The specific implementation method for determining the link weight corresponding to each link is consistent for each link, and will be explained here using any one link as an example. For a particular link, at least one dimension of link transceiver information can be extracted from the communication protocol message corresponding to that link. Furthermore, the current link weight corresponding to that link can be determined based on a pre-set weight determination function and this link transceiver information. Based on the same processing method, the current link weight corresponding to each link in the cross-domain communication network at the current moment can be obtained.

[0039] Exemplarily, the cross-domain communication network includes six communication nodes, namely, airspace communication node 1, airspace communication node 2, surface communication node 3, surface communication node 4, underwater communication node 5, and underwater communication node 6. At the current moment, the links include: link A between airspace communication node 1 and airspace communication node 2 (link A is a radio communication mode), link B between airspace communication node 2 and surface communication node 4 (link B is a radio communication mode), link C between airspace communication node 2 and underwater communication node 5 (link C is a laser communication mode), link D between surface communication node 3 and surface communication node 4 (link D is a radio communication mode), link E between surface communication node 3 and underwater communication node 5 (link E is an underwater acoustic communication mode), and link F between underwater communication node 5 and underwater communication node 6 (link F is an underwater acoustic communication mode). Furthermore, for link A, the current link weight W1 corresponding to link A can be determined based on at least one link transmission and reception information corresponding to link A and a pre-set weight determination function. Based on the same processing manner, the current link weights W2, ..., corresponding to link B and the current link weight W6 corresponding to link F can be determined.

[0040] S120: Construct a current weighted network graph corresponding to the cross-domain communication network based on each communication node, each link, and the corresponding current link weight.

[0041] The weighted network graph refers to a graph in which weights are assigned to the connection edges of the network graph. The current weighted network graph is a weighted network graph determined according to the current link weights determined at the current moment.

[0042] Specifically, each communication node in the cross-domain communication network can be used as a node, each link at the current moment as a connection edge, and the current link weight corresponding to each link as the weight value of the connection edge to construct a current weighted network graph corresponding to the cross-domain communication network.

[0043] Based on the above exemplary embodiment, the schematic diagram of the current weighted network diagram corresponding to the cross-domain communication network is shown in Figure 2 ,like Figure 2As shown, the current weighted network diagram includes 6 nodes, among which the two nodes filled with gray represent airspace communication nodes 1 and 2, the two nodes filled with black represent surface communication nodes 3 and 4, and the two nodes filled with white represent underwater communication nodes 5 and 6. The first connection between nodes 1 and 2 represents link A, and the weight value of the first connection is W1 (i.e., the current link weight corresponding to link A); the second connection between nodes 2 and 4 represents link B, and the weight value of the second connection is W2; the third connection between nodes 2 and 5 represents link C, and the weight value of the third connection is W3; the fourth connection between nodes 3 and 4 represents link D, and the weight value of the fourth connection is W4; the fifth connection between nodes 3 and 5 represents link E, and the weight value of the fifth connection is W5; the sixth connection between nodes 5 and 6 represents link F, and the weight value of the fifth connection is W6.

[0044] S130: Based on the current weighted network graph, determine the route from each communication node to other communication nodes in the cross-domain communication network at the current moment.

[0045] The route includes at least one link.

[0046] In this embodiment, the processing method for each node in the current network diagram is consistent, and any one of them is used as an example for description. Figure 2 , taking node 1 as an example, the optimal path L1 between node 1 and node 2, the optimal path L2 between node 1 and node 3, the optimal path L3 between node 1 and node 4, the optimal path L4 between node 1 and node 5, and the optimal path L5 between node 1 and node 6 can be determined by the current weighted network graph and the preset path optimization algorithm (for example, Dijkstra algorithm). According to the corresponding relationship between each node and each communication node in the cross-domain communication network in the current weighted network graph, the corresponding relationship between each connection edge and each link, and each optimal path, it can be determined that the first route between the airspace communication node 1 and the airspace communication node 2 includes the link corresponding to each connection edge in the optimal path L1, the second route between the airspace communication node 1 and the surface domain communication node 3 includes the link corresponding to each connection edge in the optimal path L2, and so on. Ultimately, the route from each communication node to each other communication node in the cross-domain communication network at the current moment can be obtained. Furthermore, these routes can be stored in a preset storage unit in the form of a routing table. Thus, when data needs to be transmitted from a source communication node to a destination communication node, the routing table can be directly called to determine the data to be transmitted based on various routes.

[0047] After step S130, the next moment of the current moment may be updated to the current moment, and steps S110-S130 may be repeatedly executed in a loop to achieve the technical effect of dynamically determining the routing table in real time.

[0048] The technical solution of the embodiment of the present invention is applied to a cross-domain communication network including multiple communication nodes in at least two spatial domains. The method includes: determining each link of the cross-domain communication network at the current moment, and determining the current link weight corresponding to each link based on the link transceiver information of each link at the current moment, wherein the link indicates that there is a direct communication relationship between two communication nodes, and then, based on each communication node, each link and the corresponding current link weight, constructing a current weighted network graph corresponding to the cross-domain communication network, thereby determining the route from each communication node to each other communication node in the cross-domain communication network at the current moment based on the current weighted network graph, wherein the route includes at least one link. The technical solution provided by the embodiment of the present invention, for such a complex communication scenario of the cross-domain communication network, can dynamically adjust the link weight in real time according to the link transceiver information of different links in the cross-domain communication network, thereby determining the route corresponding to the current moment according to the adjusted link weight, and can determine the route with the highest transmission efficiency, thereby improving the overall data transmission efficiency of the cross-domain communication network.

[0049] Example 2

[0050] Figure 3 This is a flowchart of an intelligent routing method applied to a cross-domain communication network provided in the second embodiment of the present invention. Based on the above embodiment, S110 and S130 are described in detail, wherein the technical terms that are the same as or corresponding to the above embodiment are not repeated here.

[0051] like Figure 3 As shown, the method includes:

[0052] S210 : For each link, based on the link receiving and transmitting information of the link at the current moment, determine a communication quality evaluation attribute and a link priority attribute corresponding to the link.

[0053] The communication quality assessment attribute is used to characterize the data transmission quality of the link; the link priority attribute is used to characterize the importance of the link data transmission task.

[0054] Specifically, link transmission and reception information includes: packet flow, packet transmission time, link bandwidth, link transmission delay, and link packet loss rate. Packet flow refers to the overall number of packets transmitted over a network link per unit time. It is a fundamental metric for measuring network load and forwarding performance. Packet transmission time refers to the time it takes for a packet to travel from the sender (a communication node) to the receiver (another communication node) on the link. Link bandwidth refers to the data transmission rate or capacity of a network link, typically expressed in bits per second (bps). Link transmission delay is primarily related to packet size and transmission rate, and is calculated as: Link transmission delay = packet size / transmission rate. The packet loss rate refers to the rate at which packets fail to reach their destination during network transmission. This rate reflects packet loss during transmission and is a key indicator of network performance.

[0055] Optionally, a specific implementation method of determining the communication quality evaluation attribute and the link priority attribute corresponding to the link based on the link transmission and reception information of the link at the current moment may include:

[0056] (1) Based on the link traffic information, the time required for link transmission of data packets, and the link bandwidth information, the communication quality evaluation attribute corresponding to the link is determined.

[0057] More specifically, based on the link traffic information corresponding to the link and the time required for the link to transmit a data packet, the average traffic information is determined; based on the average traffic information and the link bandwidth information, the link load information corresponding to the link is determined; based on the link load information, the communication quality evaluation attribute corresponding to the link is determined.

[0058] In this embodiment, the link traffic information corresponding to the link and the time required for the link to transmit a data packet can be divided to obtain average traffic information K(t). Furthermore, the average traffic information K(t) and the link bandwidth information can be divided to obtain link load information L(t) corresponding to the link. Thus, an exponential operation is performed on the link load information to obtain the communication quality evaluation attribute corresponding to the link. The formula for determining the communication quality evaluation attribute can be expressed as follows:

[0059] Q(t)=exp(L(t))

[0060] Where Q(t) represents the communication quality evaluation attribute, and L(t) represents the link load information.

[0061] In this embodiment, the effective bandwidth of different links can be calculated by extracting the average traffic of different links over a period of time dynamically fed back in the network, and then the information is fed back to the routing process, so as to predict possible congestion in advance and avoid the occurrence of congestion.

[0062] (2) Based on the link transmission delay, link packet loss rate and preset link level information, determine the link priority attribute corresponding to the link.

[0063] The preset link level information is a link level information determined from a plurality of predefined level information when determining a link based on the importance of the link's data transmission task. For example, the more important the link's data transmission task, the smaller the link level information value; the less important the link's data transmission task, the larger the link level information value.

[0064] More specifically, the inverse of the preset link level information is taken and an exponential operation is performed to obtain the importance level reference value corresponding to the link; a weighted sum operation is performed on the link transmission delay, link packet loss rate and importance level reference value to obtain the link priority attribute corresponding to the link.

[0065] In this embodiment, the preset link level information can be calculated according to a preset data processing method to obtain an importance level reference value. For example, the inverse of the preset link level information can be taken and then an exponential operation can be performed to obtain the importance level reference value corresponding to the link. The formula for determining the importance level reference value can be expressed as:

[0066] S(t)=exp(1 / rt)

[0067] Where S(t) represents the communication quality evaluation attribute, and rt represents the preset link level information.

[0068] Furthermore, based on the link transmission delay, link packet loss rate, and importance level reference value, the calculation formula for obtaining the link priority attribute corresponding to the link can be expressed as:

[0069] P(t)=αd(t)+βl(t)+γS(t)

[0070] Where P(t) represents the link priority attribute, d(t) represents the link transmission delay, l(t) represents the link packet loss rate, S(t) represents the importance level reference value, and α, β, and γ are pre-set reference weight values.

[0071] S220: Determine a current link weight corresponding to the link based on the communication quality evaluation attribute and the link priority attribute.

[0072] In this embodiment, a weight prediction model for determining the current link weight corresponding to a link according to the communication quality evaluation attribute and the link priority attribute may be pre-trained.

[0073] Exemplarily, a plurality of sets of simulated link transceiver information can be collected through a discrete event simulation model of a network system, and samples corresponding to each simulated link are determined based on these simulated link transceiver information, each sample including a communication quality assessment attribute and a link priority attribute corresponding to a certain link, thereby obtaining a training sample set. Furthermore, each sample data in the training sample set is normalized, and the data is divided into a training set and a test set. Subsequently, the dimension and format of the divided sample data are adjusted to the input shape required by the model, and a loss function and an optimizer are defined so that subsequent training can optimize the model parameters through back propagation. Furthermore, training of a weight prediction model to be trained is started (for example, the weight prediction model to be trained can be a package deep learning model), and the predicted link weight output by the initial weight prediction model is compared with the label weight value (the label weight value is a theoretical optimal value obtained by mathematically calculating different state information of the link through theoretical considerations), and the model parameters of the weight prediction model to be trained are continuously optimized until the model training termination condition is met, and a trained weight prediction model is obtained, and the weight prediction model is saved.

[0074] In specific applications, the communication quality assessment attributes and link priority attributes corresponding to the link are input into a pre-trained weight prediction model, and the weight prediction model can output the current link weight value corresponding to the link.

[0075] S230: Based on each communication node, each link and the corresponding current link weight, construct a current weighted network graph corresponding to the cross-domain communication network.

[0076] S240 . For each node in the current weighted network graph, determine each node except the first node as a second node.

[0077] In this embodiment, see Figure 2 ,like Figure 2 As shown, node 1 can be regarded as the first node, and node 2, node 3, node 4, node 5 and node 6 can be regarded as the second nodes.

[0078] S250: Taking the first node as the starting node and any second node as the ending node, a preset path optimization algorithm is used to determine the optimal path between the first node and the second node.

[0079] Based on the above example, node 1 can be used as the starting node and node 2 as the ending node, and a preset path optimization algorithm (for example, Dijkstra algorithm) can be used to determine the optimal path L1 between node 1 and node 2. Based on the same processing method, the optimal path L2 between node 1 and node 3, the optimal path L3 between node 1 and node 4, the optimal path L4 between node 1 and node 5, and the optimal path L5 between node 1 and node 6 can be determined.

[0080] Next, with node 2 as the first node and node 1, node 2, node 4, node 5 and node 6 as the second node, S240-S250 may be repeatedly executed until the optimal path between each node and each node except itself is determined.

[0081] S260. Determine a first communication node corresponding to the first node, a target communication node corresponding to the second node, and target links corresponding to the optimal path.

[0082] In this embodiment, based on the correspondence between each node and each communication node and the correspondence between each connection edge and each link in the current weighted network graph, it is easy to determine the first communication node corresponding to the first node, the target communication node corresponding to the second node, and each target link corresponding to the optimal path.

[0083] S270: Determine the path formed by each target link as the route from the first communication node to the target communication node in the cross-domain communication network at the current moment.

[0084] In this embodiment, the route from the first communication node to the target communication node in the cross-domain communication network at the current moment is a path formed by sequentially connecting the target links.

[0085] The technical solution of the embodiment of the present invention determines, for each link, the communication quality evaluation attribute and the link priority attribute corresponding to the link based on the link receiving and sending information of the link at the current moment, and then, based on the communication quality evaluation attribute and the link priority attribute, determines the current link weight corresponding to the link, and then constructs the current weighted network graph corresponding to the cross-domain communication network based on each communication node, each link and the corresponding current link weight. Furthermore, for each node in the current weighted network graph, each node except the first node is determined as the second node, and the first node is used as the starting node and any second node is used as the ending node. A preset path optimization algorithm is used to determine the optimal path between the first node and the second node, and then, the first communication node corresponding to the first node, the target communication node corresponding to the second node, and each target link corresponding to the optimal path are determined. Thus, the path formed by each target link is determined as the route from the first communication node to the target communication node in the cross-domain communication network at the current moment. The technical solution provided by the embodiment of the present invention is aimed at such complex communication scenarios of cross-domain communication networks. It can dynamically adjust the link weights in real time based on the link receiving and transmitting information of different links in the cross-domain communication network, thereby determining the route corresponding to the current moment based on the adjusted link weights, and determining the route with the highest transmission efficiency. When adjusting the link weights, the effective bandwidth of different links is calculated by extracting the average traffic of different links dynamically fed back in the network over a period of time, and then feeding this information back to the routing process, thereby predicting possible congestion in advance, avoiding the occurrence of congestion, and further improving the overall data transmission efficiency of the cross-domain communication network.

[0086] Example 3

[0087] Next, a specific example is used to illustrate the specific implementation process of the intelligent routing method applied to the cross-domain communication network. In this embodiment, a simulated cross-domain communication network is constructed by a discrete event simulation model of the network system to simulate the link transmission and reception information generated in the actual cross-domain communication network. Figure 4 This is a general flow chart of an intelligent routing method applied to a cross-domain communication network according to Embodiment 3 of the present invention, which mainly includes the following steps:

[0088] Step 1. Extract the link status information (i.e., link transmission and reception information) corresponding to each link at the current moment from the simulated cross-domain communication network, mainly including: data packet flow information f flowing through the link, the time required for the link to transmit the data packet t, the link bandwidth information bwa, the link transmission delay dt, and the link packet loss rate lt, and obtain the preset link level information rt corresponding to each link.

[0089] Step 2: Determine the average flow information based on the link traffic information corresponding to the link and the time required for the link to transmit data packets. Then, based on the average flow information and the link bandwidth information, determine the link load information corresponding to the link. Furthermore, based on the link load information, determine the communication quality evaluation attribute Q(t) corresponding to the link.

[0090] Step 3: Based on the link transmission delay, link packet loss rate and preset link level information, determine the link priority attribute P(t) corresponding to the link.

[0091] Step 4: Input the communication quality assessment attribute Q(t) and link priority attribute P(t) corresponding to each link into the pre-trained intelligent routing model (i.e., weight prediction model) to obtain the effective weight corresponding to each link (i.e., current link weight).

[0092] Step 5: Generate the current weighted network graph based on the generated effective weights of each link and the network information. This is achieved by: each communication node constructing a complete network topology based on the received link state information. This topology includes information about all communication nodes and links. Based on this constructed network topology, the communication node generates a weighted network graph, where nodes represent communication nodes, edges represent connected links, and edge weights are the effective weights calculated in the above steps.

[0093] Step 6: Update the current weighted network graph to the Dijkstra algorithm, use the Dijkstra algorithm to find the best routing path for each node in the entire network, and store all the best routing paths in the routing table.

[0094] Step 7 periodically repeats Step 1-Step 6 to ensure the timeliness of the algorithm.

[0095] In this embodiment, after data communication is performed according to a route in the routing table, the communication quality of the route can be evaluated based on the throughput and packet loss rate during the communication process. Throughput refers to the amount of data successfully transmitted per unit time by a network, device, port, virtual circuit, or other facility, and is typically measured in units such as bits, bytes, or packets. The packet loss rate is calculated by dividing the number of lost packets by the number of sent packets.

[0096] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0097] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. An intelligent routing method applied to a cross-domain communication network, characterized in that: The cross-domain communication network includes multiple communication nodes in at least two spatial domains, including: Determine each link of the cross-domain communication network at the current moment, and determine a current link weight corresponding to each link based on link transceiver information of each link at the current moment; wherein the link indicates that there is a direct communication relationship between two communication nodes; Constructing a current weighted network graph corresponding to the cross-domain communication network based on each of the communication nodes, each of the links, and the corresponding current link weights; Based on the current weighted network graph, determining a route from each communication node to each other communication node in the cross-domain communication network at the current moment; wherein the route includes at least one link; The link transceiver information includes: data packet flow information flowing through the link, time required for link data packet transmission, link bandwidth information, link transmission delay, and link packet loss rate. The current link weight corresponding to each link is determined based on the link transceiver information of each link at the current moment, including: For each of the links, determining a communication quality assessment attribute corresponding to the link based on the link traffic information corresponding to the link, the time required for the link to transmit a data packet, and the link bandwidth information; Determining a link priority attribute corresponding to the link based on the link transmission delay, the link packet loss rate, and preset link level information; The communication quality evaluation attribute and the link priority attribute corresponding to the link are input into a pre-trained weight prediction model to obtain a current link weight value corresponding to the link.

2. The method according to claim 1, characterized in that The determining of a communication quality evaluation attribute corresponding to the link based on the link traffic information corresponding to the link, the time required for the link to transmit a data packet, and the link bandwidth information includes: Determining average flow information based on the link flow information corresponding to the link and the time required for the link to transmit a data packet; Determining link load information corresponding to the link based on the average traffic information and the link bandwidth information; Based on the link load information, a communication quality evaluation attribute corresponding to the link is determined.

3. The method according to claim 2, characterized in that The determining, based on the link load information, a communication quality assessment attribute corresponding to the link includes: An exponential operation is performed on the link load information to obtain a communication quality evaluation attribute corresponding to the link.

4. The method according to claim 1, wherein The determining, based on the link transmission delay, the link packet loss rate, and the preset link level information, a link priority attribute corresponding to the link includes: Taking the inverse of the preset link level information and performing an exponential operation thereon to obtain an importance level reference value corresponding to the link; A weighted sum operation is performed on the link transmission delay, the link packet loss rate, and the importance level reference value to obtain a link priority attribute corresponding to the link.

5. The method according to claim 1, wherein The constructing a current weighted network graph corresponding to the cross-domain communication network based on each of the communication nodes, each of the links, and the corresponding current link weights includes: A current weighted network graph corresponding to the cross-domain communication network is constructed by taking each of the communication nodes in the cross-domain communication network as a node, each of the links as a connection edge, and the current link weight corresponding to each of the links as the weight value of the corresponding connection edge.

6. The method according to claim 1, wherein The determining, based on the current weighted network graph, a route from each communication node to each other communication node in the cross-domain communication network at a current moment, includes: For each of the nodes in the current weighted network graph, determining each node except the first node as a second node; Taking the first node as a starting node and any of the second nodes as an ending node, and using a preset path optimization algorithm, determining an optimal path between the first node and the second node; Determining a first communication node corresponding to the first node, a target communication node corresponding to the second node, and target links corresponding to the optimal path; The path formed by each of the target links is determined as the route from the first communication node to the target communication node in the cross-domain communication network at the current moment.

7. The method according to claim 1, characterized in that The at least two spatial domains include an airspace domain, a surface domain, and an underwater domain, and the multiple communication nodes include at least one airspace communication node, at least one surface domain communication node, and at least one underwater domain communication node; The link between the airspace communication node and the airspace communication node corresponds to the radio communication mode; the link between the airspace communication node and the surface domain communication node corresponds to the radio communication mode; the link between the airspace communication node and the underwater domain communication node corresponds to the laser communication mode; the link between the surface domain communication node and the surface domain communication node corresponds to the radio communication mode; the link between the surface domain communication node and the underwater domain communication node corresponds to the underwater acoustic communication mode; the link between the underwater domain communication node and the underwater domain communication node corresponds to the underwater acoustic communication mode.

Images